[sr-dev] crash (segfault) on Kamailio master branch

Jason Penton jason.penton at gmail.com
Mon Oct 5 08:29:30 CEST 2015


Hey guys,

Not sure if there have been any changes but I have an interesting problem
here when using TCP:

*The BT is as follows:*

#0  local_timer_list_expire (h=0xa0f128 <tcp_reader_ltimer+178664>,
t=723807134, l=0x9e3740 <tcp_reader_ltimer>) at local_timer.c:198
#1  local_timer_expire (t=723807134, h=<optimized out>) at local_timer.c:227
#2  local_timer_run (lt=lt at entry=0x9e3740 <tcp_reader_ltimer>,
saved_ticks=723807150) at local_timer.c:250
#3  0x00000000005d8417 in tcp_reader_timer_run () at tcp_read.c:1682
#4  tcp_receive_loop (unix_sock=<optimized out>) at tcp_read.c:1734
#5  0x00000000005c81c8 in tcp_init_children () at tcp_main.c:4788
#6  0x00000000004a9da3 in main_loop () at main.c:1664
#7  0x000000000042411e in main (argc=<optimized out>, argv=<optimized out>)
at main.c:2566

This seems to be related to clearing timers for TCP connections. The crash
is related to the following code:

*_timer_rm_list(tl)*

where it does a null pointer deref on tl->next and tl->prev, which,
according to the bt, are null (see below).

*(gdb) print *tl*
$14 = {next = 0x0, prev = 0x0, expire = 723807134, initial_timeout = 32,
data = 0x7fbbb05aa628, f = 0x5d02f0 <tcpconn_read_timeout>, flags = 512,
slow_idx = 0}

Any ideas?

Cheers
Jason
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.sip-router.org/pipermail/sr-dev/attachments/20151005/64b984c7/attachment-0001.html>


More information about the sr-dev mailing list